Project Summary/Abstract The goal of the proposed research is to characterize the location and functional interaction over time of areas of the brain involved in aspects of sentence comprehension. The research is undertaken within a framework that sees comprehension as having multiple information sources, and as being interactive and situated. This view of sentence comprehension maintains that the propositional meaning of a sentence is, in the final analysis, determined by its syntactic structure and lexical content, and, when comprehension is successful, this meaning is arrived at by a process that assigns syntactic structure (parsing) and combines it with lexical meanings to determine propositional meaning (syntactically-based interpretation, which we will refer to simply as interpretation). However, these integrated, central processes are not the only sources of propositional representations that are considered as candidates for the meaning of a sentence. Propositional meanings are also assigned by asyntactic processes such as heuristics that assign agent to a preverbal noun (comprehension has multiple sources). These multiple sources yield candidates for the meaning of a sentence on-line, which are eventually eliminated in favor of the syntactically licensed interpretation(s) of a sentence. In addition, parsing and interpretation are affected in an on-line fashion by a variety of types of information such as the frequency of lexical forms and their co-occurrence (comprehension is interactive), and by non- linguistic factors such as task conditions (comprehension is situated). The proposed research will identify areas of the brain that are involved in aspects of parsing and interpretive operations and their on-line interaction with non-linguistic processes. We will study the two major types of these interactions - those between parsing and interpretation and perceptual and motor processes and those between parsing and interpretation and semantic memory. We will use MR-constrained integrated magnetoencephalography (MEG) and electroelcephalography (EEG) to obtain measures of neural activity that have both fine-grained temporal and spatial resolution, and apply Granger causation analyses (Granger, 1969) to these data to identify patterns of directional causal influence between functionally interpretable ROIs during sentence processing.